Bone mineral accrual is maximum in adolescence and 90% of peak bone mass is achieved by 18 y of age. Insults suffered in adolescence can thus result in permanent deficits with increasd fracture risk in later life. Anorexia nervosa (AN) is associated with decreased bone formation and signifiant osteopenia and osteoporosis. Osteopenia is also more severe when AN begins in adolescence than in adult life. The pathogenesis of bone loss, however, is not well understood and no treatment has so far proven effective in treating low bone density (BMD) in adolescents with AN. Some improvement in BMD occurs with weight recovery, but a third of AN subjects remain osteoporotic. Calcium and vitamin D supplementation does not improve BMD in AN. Adults with AN have have high growth hormone (GH) and low IGF-I levels, suggestive of a nutritionally acquired 'GH resistance'. GH, both directly and through IGF-I, plays an important role in osteoblast differentiation. However, the presence of GH resistance in adolescents with AN in different pubertal stages, and the contribution of this 'resistance' to low BMD has not been examined. AN is associated with very low estrogen levels, but despite the known inhibitory effect of estrogen on bone resorption, oral estrogen was not effective in improving bone density in adults with AN. Oral estrogen suppresses IGF-I, and this may prevent the improvement in bone mass expected from anti-resorptioe effects of estrogen. Conversely, transdermal estrogen does not lower IGF-I, and the effects of transdermally administered estrogen on bone turnover in AN have not been examined. One aim of the proposal is to determine whether adolescents with AN are GH resistant, and whether this 'resistance' contributes to decreased bone formation and low bone density in AN. To investigate this hypothesis, we will examine GH secretory patterns in AN and matched controls, and determine if alterations in GH secretion predict changes in whole body metabolism, body composition, bone formation markers and BMD. We also hypothesize that administration of transdermal estrogen to mature adolescent girls with AN will result in decreased bone loss and increased bone formation (by increasing IGF-I levels). To investigate this hypothesis, we will examine bone metabolism in adolescent girls with AN before and six months after administration of transdermal estrogen. The effects of estrogen in weight recovered vs. non-weight recovered AN will also be studied.